Impact of Venous Thromboembolism on Readmissions and Resource Use Following Emergency General Surgery

Abstract

Background

Acute deep vein thrombosis and pulmonary embolism collectively known as venous thromboembolism (VTE), are associated with increased risk of poor clinical sequelae during inpatient hospitalizations. We examined the association of VTE with mortality, readmissions, and costs among patients undergoing emergency general surgery (EGS) operations using a national cohort.

Methods

Adult hospitalizations for EGS (laparotomy, small bowel resection, large bowel resection, appendectomy, lysis of adhesions, cholecystectomy, and repair of perforated ulcer) within two days of admission were identified in the 2016-18 Nationwide Readmissions Database. Hospitalizations were stratified based on diagnosis of VTE and others (n-VTE).

Results

Of an estimated 860,747 EGS patients 7,513, (.87%) developed VTE during the index hospitalization. Patients in the VTE group were on average older (65.5 ± 15.3 vs 54.8 ± 18.6 years, P < .001) and more commonly male (46.7 vs 39.3%, P < .001). Venous thromboembolism was independently associated with greater odds of mortality (AOR:1.7 95% CI 1.6-1.9), increased costs (+27,700 95% CI 23, 100-28,300) and greater odds of 30-day readmissions (AOR 1.3 95% CI 1.2-1.4).

Discussion

Despite national efforts to reduce its incidence, VTE affects nearly 1/100 EGS patients and is associated with increased odds of mortality as well as costs, and readmissions. Tailored approaches are warranted to reduce the impact of this pernicious complication.

Keywords

emergency general surgery general surgery acute care surgery

Introduction

Emergency general surgery (EGS) represents approximately 12% of operations performed annually in the US, accounting for 7% of all hospitalizations.^1,2 However, this group of operations leads to nearly 30% of all postoperative complications as well as >50% of surgical deaths.³ Several large series have demonstrated EGS to be an independent risk factor for postoperative complications.^1,4 Due to limited opportunity for preoperative risk modification in EGS, recent efforts have focused on management of specific postoperative adverse events including venous thromboembolism.^1,2

With an annual economic burden of $10 billion in the US alone, venous thromboembolism (VTE) represents a major potentially preventable source of morbidity and associated death among hospitalized patients.⁵ Despite efforts to standardize prophylaxis guidelines, VTE is frequently encountered in EGS patients.^1,3,4 The present study examined the prevalence and risk factors for VTE across the spectrum of EGS operations in a national cohort. We further characterized the incidence of VTE upon readmission within 30 days of index discharge. We hypothesized development of VTE to be associated with inferior clinical outcomes and increased resource use following EGS.

Methods

The 2016-2018 Nationwide Readmissions Database (NRD) was used to identify all non-elective adult hospitalizations (≥18 years) using International Classification of Diseases 10^th Revisions (ICD-10) codes corresponding to any of the following EGS operations: large bowel resection, small bowel resection, cholecystectomy, repair of perforated peptic ulcer, lysis of adhesions, and appendectomy within 3 days of admission as previously described.⁶ The NRD is the largest, all-payer readmissions database that collects data from 28 participating states and allows tor tracking of readmissions within each calendar year.⁷ Patients with diagnosis of chronic VTE or history of inferior vena cava filter placement (1.4%), as well as those with missing data or indication for trauma were excluded (9.8%). The EGS cohort was stratified into VTE and n-VTE based on diagnosis of VTE during index hospitalization. Variables of interest included sex, age, insurance status, household income quartile, and hospital bed size, as defined by the NRD data dictionary.⁷ Major adverse events (MAE) were identified by presence of any of the following ICD-10 derived complications: cardiac, respiratory, renal, cerebrovascular, infection, and cerebrovascular. The van Walraven modification of the Elixhauser Comorbidity Index, a previously validated composite index of 30 chronic comorbidities, was used to quantify the overall burden of disease.⁸ Costs were calculated using specific cost-to-charge ratios and inflation adjusted for 2018 using the Personal Healthcare Price Index.

The primary outcome of interest was the development of VTE during index hospitalization. Other study endpoints included in-hospital mortality, index hospitalization costs, MAE, and all-cause, non-elective 30-day readmission. Categorical variables are reported as frequency and compared using Pearson’s χ² tests. Continuous variables are reported as means with standard deviation and compared using the adjusted Wald test. Multivariable regressions were developed to test association of covariates with outcomes of interest (Table 1). Regression outputs are reported as adjusted odds ratios (AOR) or beta-coefficients (β) with 95% confidence intervals (95% CI), as appropriate. Statistical significance was set at α < .05. This study was deemed exempt from full review by the institutional review board at the University of California, Los Angeles. All statistical analysis was performed using Stata 16.1 (Statacorp, College Station, Texas).

Table 1.

Patient Demographics Across VTE and n-VTE Groups.

	n-VTE (n = 853 234)	VTE (n = 7513)	P- value
Female (%)	60.7	53.3	<.001
Age (mean, SD)	54.8 ± 18.6	65.5 ± 15.3	<.001
Elixhauser index	2.3 ± 2.0	4.9 ± 2.2
Operative characteristics (%)			<.001
Repair of perforated duodenal ulcer	96.5	3.5
Large bowel resection	96.9	3.1
Small bowel resection	97.5	2.5
Lysis of adhesions	98.7	1.3
Appendectomy	99.3	0.7
Cholecystectomy	99.7	0.3
Payer status (%)			<.001
Medicare	37.0	59.3
Privately insured	36.1	23.4
Medicaid	17.6	11.6
Other	9.4	5.7
Comorbidities (%)
Hypertension	46.6	58.2	<.001
Diabetes	18.7	24.1	<.001
Chronic lung disease	14.6	21.3	<.001
Arrhythmia	13.5	21.3	<.001
Hospital bed size (%)			<.001
Large bed size	53.4	60.7
Medium bed size	29.1	25.7
Small bed size	17.5	13.6

Results

Of an estimated 860,747 and 7513 EGS hospitalizations meeting criteria, (.87%) comprised the VTE cohort. Compared to others, VTE patients were older (65.5 ± 15.3 vs 54.8 ± 18.6, P < .001) and more frequently male (60.7% vs 53.3, P < .001). Additionally, VTE had a greater mean Elixhauser Comorbidity Index (4.9 ± 2.2, vs 2.3 ± 2.0, P < .001). Stratification by operative procedure demonstrated variability in the prevalence of VTE with the highest rates for repair of perforated ulcer (3.5%) and the lowest for cholecystectomy (.3%) (P < .001). Unadjusted comparisons of outcomes between groups are reported in Table 2. Compared to n-VTE, those in the VTE cohort faced significantly higher rates of mortality as well as non-home discharge and increased duration of stay. Compared to others, VTE patients had increased rates of MAE (69.9% vs 16.2%, P < .001).

Table 2.

Unadjusted Outcomes Between VTE and n-VTE Groups. *MAE: Major Adverse Events.

	n-VTE (n = 853 234)	VTE (n = 7513)	P- value
Mortality (%)	1.8	12.3	<.001
Complications (%)
MAE*	16.2	69.9	<.001
Renal	10.7	42.2	<.001
Respiratory	8.3	55.1	<.001
Infectious	2.9	20.3	<.001
Cardiac	2.5	15.5	<.001
Stroke	.03	.27	<.001
Non-home discharge (%)	40.8	44.4	<.001
Costs in $1000 (mean, SD)	11.8 ± 19.3	45.5 ± 72.0	<.001
Length of stay in days (mean, SD)	4.8 ± 5.9	19.3 ± 19.5	<.001
Readmissions (%)
30-day readmission	8.6	18.7	<.001
30-day non-elective readmission	7.4	16.3	<.001

After risk adjustment, patients undergoing repair of perforated ulcer (AOR: 5.5 95% CI 4.9-6.3, Ref: cholecystectomy) and large bowel resection (AOR: 4.5 95% CI 4.1-4.9, Ref: cholecystectomy) had the greatest odds of developing VTE. Other factors associated with increased odds of VTE included male sex (AOR: 1.2 95% CI 1.1-1.3) and increasing age (AOR: 1.01-year 95% CI 1.00-1.02). The presence of VTE was independently associated with greater odds of mortality (AOR:1.7 95% CI 1.6-1.9) (Table 3), MAE (AOR: 3.4 95% CI 3.1-3.6), and 30-day readmissions (AOR: 1.3 95% CI 1.2-1.4). Similarly, VTE was associated with incremental increases in hospitalization costs by +27 700 (95% CI 23 100 - 28 300) and LOS +7.7 days (95% CI 6.6-8.8).

Table 3.

Association of VTE with Select Outcomes *ROC: Receiver-Operator Characteristics.

	VTE	ROC*
Mortality	1.7 (1.6-1.9)	.93
Non-home discharge	3.6 (3.3-3.9)	.88
Complications
Adverse event	3.4 (3.1-3.7)	.87
Infectious	2.3 (1.9-2.6)	.87
Acute kidney injury	1.6 (1.5-1.8)	.85
Respiratory	4.0 (3.7-4.4)	.87
Cardiac	2.3 (2-2.6)	.94
Cerebrovascular	2.4 (2.0-3.0)	.88
30 readmissions day	1.3 (1.2-1.4)	.69
Hospitalization costs$*	+27,700 (23,100-28,300)	—

Risk-adjusted rates of mortality were persistently higher in the presence of VTE across all operations (Figure 1), with the greatest risk-adjusted rates among those undergoing repair of perforated ulcer (18.2% vs 9.4%, P < .001). Similarly, VTE patients undergoing repair of perforated ulcer had the greatest risk-adjusted rate of MAE (82.8% vs 45.6%, P < .001), and cholecystectomy (52.8% vs 9.9%, P < .001) the lowest. Compared to others, VTE cohort exhibited significantly increased risk-adjusted rates of 30-day readmissions in patients undergoing lysis of adhesions (25.3% vs 12.0%) and large bowel resection (23.5% vs 16.9%, P < .001 for all). VTE patients undergoing repair of perforated ulcer incurred greater costs ($82,400 ± 5400) compared to n-VTE ($36,100 ± 600, P < .001).

Discussion

In the present study of EGS patients, we found VTE to occur in approximately .9% of hospitalizations. Presence of VTE was associated with a nearly 2-fold increased odds of in-hospital death and greater inpatient expenditures. Notably, VTE exhibited a more profound impact on outcomes following complex EGS operations, such as repair of perforated duodenal ulcer. Several of these findings merit further discussion.

Major abdominal surgery has been widely recognized as a risk-factor for VTE, in part due to the inflammatory state and disturbance of the coagulation cascade.^9,10 Surgical candidates are often predisposed to prolonged states of immobility, further increasing risk of venous thrombosis.^1,10-12 Similar to other work, we found male sex to be associated within increased odds of developing VTE and further investigation is required to understand the mechanism of this phenomenon.^1,9 Significant variation was observed in the prevalence of VTE between EGS subtypes, with large bowel resection associated with the highest rates. Prior work has shown increased likelihood of VTE in a wide range of open operations and surgical modality to be one of the greatest predictors of inferior outcomes.^1,12,13 As risk modification prior to emergency surgical intervention is generally not feasible, implementation of strategies for prompt recognition of post-operative complications may enhance outcomes.^1,2

Venous thromboembolism was associated with near doubling of mortality across EGS, consistent with previous studies.^1,12,14 Our data represents a contemporary cohort and suggests that VTE continues to be a major driver of surgical morbidity. Instigating factors such as the presence of central venous catheters, delays in starting VTE prophylaxis, and the lack of ambulation likely contribute to greater risk of VTE following complex operations.^1,13 Whether this relationship is truly attributable to VTE or represents other coexisting conditions, merits further investigation.

Existing literature has shown VTE to result in increased costs ranging from $12,000 to $15,000 (adjusted to 2014) among medically and surgically hospitalizations.^14,15 In our cohort of EGS patients, we found a risk-adjusted increase of over $25,000 in increased costs, and an increase in stay by nearly 1 week. Nearly one-fifth of our study population with VTE was found to be readmitted within 30 days.¹⁴ The high rate of readmission demonstrates the burden of VTE to extend beyond acute hospitalization, with prior work attributing the projected overall costs of hospital-related VTE to range from $7-10 billion in excess expenditures annually.^13,14

The present study has several limitations. Since outpatient diagnoses were not ascertained, our study may have underestimated rates of VTE. Limited information is available regarding the timing and severity of VTE. Variability in adherence to pharmacologic and mechanical VTE prophylaxis was not captured in the NRD. Surgical approach (open vs laparoscopic) was not considered in the present work and may affect risk of VTE.

In conclusion, VTE is a serious complication following EGS that is associated with inferior clinical and financial outcomes. Further focused strategies are warranted to reduce the effects of this insidious complication following emergency general surgery operations.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ross

Kuhlenschmidt

Kubasiak

, et al. Association of the risk of a venous thromboembolic event in emergency vs elective general surgery. JAMA Surg. 2020;155:503-511.

McCoy

Englum

Keenan

Vaslef

Shapiro

Scarborough

. Impact of specific postoperative complications on the outcomes of emergency general surgery patients. J Trauma Acute Care Surg. 2015;78(5):912-919. discussion 918-919.

Smith

Hussain

Xiao

, et al. The importance of improving the quality of emergency surgery for a regional quality collaborative. Ann Surg. 2013;257(4):596-602.

Havens

Peetz

, et al. The excess morbidity and mortality of emergency general surgery. J Trauma Acute Care Surg. 2015;78(2):306-311.

The surgeon general’s call to action to prevent deep vein thrombosis and pulmonary embolism. Circulation. 2020;119(15): 17–22.

Hadaya

Sanaiha

Juillard

Benharash

. Impact of frailty on clinical outcomes and resource use following emergency general surgery in the United States. PLoS One. 2021;16(7):e0255122.

NRD description of data elements. https://www.hcup-us.ahrq.gov/db/nation/nrd/nrddde.jsp. (Accessed September 30, 2021).

Van Walraven

Austin

Jennings

Quan

Forster

. A modification of the elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care. 2009;47(6):626-633. doi:10.1097/MLR.0b013e31819432e5

Lyman

Carrier

Di Nisio

Hicks

Khorana

, et al. American society of hematology 2021 guidelines for management of venous thromboembolism: prevention and treatment in patients with cancer. Blood Adv. 2021;5(4):927-974.

10.

McCormack

Harrisingh

Horner

Bewley

. Venous thromboembolism in adults: Summary of updated NICE guidance on diagnosis, management, and thrombophilia testing. BMJ. 2020;369:m1565.

11.

Murphy

Vogt

Lau

, et al. Venous thromboembolism prevention in emergency general surgery: A review. JAMA Surg. 2018;153(5):479-486.

12.

Rattan

Cioci

Urréchaga

, et al. Readmission for venous thromboembolism after emergency general surgery is underreported and influenced by insurance status. J Trauma Acute Care Surg. 2021;90(1):64-72.

13.

DeWane

Davis

Schuster

Maung

Becher

. Venous thromboembolism-related readmission in emergency general surgery patients: A role for prophylaxis on discharge? J Am Coll Surg. 2018;226(6):1072-1077.e3.

14.

Henke

Kahn

Pannucci

, et al. Call to action to prevent venous thromboembolism in hospitalized patients: A policy statement from the American heart association. Circulation. 2020;141(24):e914-e931.

15.

Secemksy

Rosenfield

Kennedy

Jaff

. High burden of 30‐Day readmissions after acute venous thromboembolism in the United States. JAHA. 2022;7(13):e009047.